mlsanityData to 7 Oct 2026

Distributionally Robust Losses for Latent Covariate Mixtures

arXiv 2007.13982 ↗First posted 28 Jul 2020cs.LG stat.MLcited by 57

Not on the map. Why?

Abstract (arXiv)

While modern large-scale datasets often consist of heterogeneous subpopulations -- for example, multiple demographic groups or multiple text corpora -- the standard practice of minimizing average loss fails to guarantee uniformly low losses across all subpopulations. We propose a convex procedure that controls the worst-case performance over all subpopulations of a given size. Our procedure comes with finite-sample (nonparametric) convergence guarantees on the worst-off subpopulation. Empirically, we observe on lexical similarity, wine quality, and recidivism prediction tasks that our worst-case procedure learns models that do well against unseen subpopulations.

Key facts. Cited by 57 papers in the map (data through 7 Oct 2026). First posted on arXiv 28 Jul 2020.

Builds on

No reference list yet.

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2021: 13; 2022: 12; 2023: 11; 2024: 11; 2025: 8; 2026: 2.

57 papers.

Sort:

By area:

  1. 2021Towards Out-Of-Distribution Generalization: A Surveycited by 431
  2. 2022NICO++: Towards Better Benchmarking for Domain Generalizationcited by 86
  3. 2021The Spotlight: A General Method for Discovering Systematic Errors in Deep Learning Modelscited by 63
  4. 2021Unsupervised Learning of Debiased Representations with Pseudo-Attributescited by 44
  5. 2022Perfectly Balanced: Improving Transfer and Robustness of Supervised Contrastive Learningcited by 40
  6. 2022Pareto Invariant Risk Minimization: Towards Mitigating the Optimization Dilemma in Out-of-Distribution Generalizationcited by 38
  7. 2023Diagnosing Model Performance Under Distribution Shiftcited by 37
  8. 2021An Online Learning Approach to Interpolation and Extrapolation in Domain Generalizationcited by 30
  9. 2021Simple and near-optimal algorithms for hidden stratification and multi-group learningcited by 26
  10. 2021Regularizing towards Causal Invariance: Linear Models with Proxiescited by 20
  11. 2021A Theoretical Analysis on Independence-driven Importance Weighting for Covariate-shift Generalizationcited by 18
  12. 2022Distributionally Robust Models with Parametric Likelihood Ratioscited by 17
38 more
  1. 2022Subgroup Robustness Grows On Trees: An Empirical Baseline Investigationcited by 17
  2. 2021Preventing dataset shift from breaking machine-learning biomarkerscited by 14
  3. 2022Distributionally Robust Survival Analysis: A Novel Fairness Loss Without Demographicscited by 14
  4. 2021Algorithmic Bias and Data Bias: Understanding the Relation between Distributionally Robust Optimization and Data Curationcited by 13
  5. 2024Fairness in Survival Analysis with Distributionally Robust Optimizationcited by 13
  6. 2021A comparison of approaches to improve worst-case predictive model performance over patient subpopulationscited by 12
  7. 2022Learning from a Biased Samplecited by 12
  8. 2023Bridging Distributionally Robust Learning and Offline RL: An Approach to Mitigate Distribution Shift and Partial Data Coveragecited by 12
  9. 2024Addressing Polarization and Unfairness in Performative Predictioncited by 11
  10. 2025Task Generalization With AutoRegressive Compositional Structure: Can Learning From D Tasks Generalize to DT Tasks?cited by 11
  11. 2022Evaluating Robustness to Dataset Shift via Parametric Robustness Setscited by 10
  12. 2022Understanding the effect of sparsity on neural networks robustnesscited by 8
  13. 2024ExMap: Leveraging Explainability Heatmaps for Unsupervised Group Robustness to Spurious Correlationscited by 8
  14. 2021Learning Domain Invariant Representations by Joint Wasserstein Distance Minimizationcited by 6
  15. 2023Scaff-PD: Communication Efficient Fair and Robust Federated Learningcited by 6
  16. 2023Is this model reliable for everyone? Testing for strong calibrationcited by 6
  17. 2023Causality-oriented robustness: exploiting general noise interventionscited by 5
  18. 2023Geometry-Calibrated DRO: Combating Over-Pessimism with Free Energy Implicationscited by 5
  19. 2024Evaluating Model Performance Under Worst-case Subpopulationscited by 5
  20. 2024DRPruning: Efficient Large Language Model Pruning through Distributionally Robust Optimizationcited by 5
  21. 2025DRO: A Python Library for Distributionally Robust Optimization in Machine Learningcited by 5
  22. 2022Minimax Optimal Estimation of Stability Under Distribution Shiftcited by 4
  23. 2023Optimizer's Information Criterion: Dissecting and Correcting Bias in Data-Driven Optimizationcited by 4
  24. 2025Sample-Efficient Omniprediction for Proper Lossescited by 4
  25. 2022Long Term Fairness for Minority Groups via Performative Distributionally Robust Optimizationcited by 3
  26. 2022How does overparametrization affect performance on minority groups?cited by 2
  27. 2024Federated Fairness without Access to Sensitive Groupscited by 2
  28. 2024Contextual Optimization under Covariate Shift: A Robust Approach by Intersecting Wasserstein Ballscited by 2
  29. 2024CherryRec: Enhancing News Recommendation Quality via LLM-driven Frameworkcited by 2
  30. 2024Minimax rates of convergence for nonparametric regression under adversarial attackscited by 2
  31. 2023Predictive Heterogeneity: Measures and Applicationscited by 1
  32. 2024Footprints of Data in a Classifier: Understanding the Privacy Risks and Solution Strategiescited by 1
  33. 2024On the KL-Divergence-based Robust Satisficing Modelcited by 1
  34. 2025CTC-DRO: Robust Optimization for Reducing Language Disparities in Speech Recognitioncited by 1
  35. 2025DivIL: Unveiling and Addressing Over-Invariance for Out-of- Distribution Generalizationcited by 1
  36. 2025Distributionally Robust Federated Learning with Outlier Resiliencecited by 1
  37. 2021A Topological-Framework to Improve Analysis of Machine Learning Model Performancecited by 0
  38. 2021Learning Neural Models for Natural Language Processing in the Face of Distributional Shiftcited by 0

50 most cited of 57 All 57 (JSON)